. 1
( 13)


This page intentionally left blank
Autonomy and the Challenges to Liberalism
New Essays

In recent years, the concepts of individual autonomy and political lib-
eralism have been the subject of intense debate, but these discussions
have occurred largely within separate academic disciplines. Autonomy
and the Challenges to Liberalism contains for the ¬rst time new essays
devoted to foundational questions concerning both the notion of the
autonomous self and the nature and justi¬cation of liberalism.
Written by leading ¬gures in moral, legal, and political theory, this
volume covers, among other things, the following topics: the nature
of the self and its relation to autonomy, the social dimensions of
autonomy and the political dynamics of respect and recognition, and
the concept of autonomy underlying the principles of liberalism.

John Christman is Associate Professor of Philosophy and Political
Science at Pennsylvania State University.

Joel Anderson is Research Lecturer in the Department of Philosophy
at the University of Utrecht, The Netherlands.
Autonomy and the Challenges
to Liberalism
New Essays

Edited by
Pennsylvania State University

University of Utrecht
cambridge university press
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo

Cambridge University Press
The Edinburgh Building, Cambridge cb2 2ru, UK
Published in the United States of America by Cambridge University Press, New York
Information on this title: www.cambridge.org/9780521839518

© John Christman and Joel Anderson 2005

This book is in copyright. Subject to statutory exception and to the provision of
relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.

First published in print format 2005

isbn-13 978-0-511-10949-2 eBook (EBL)
isbn-10 0-511-10949-0 eBook (EBL)

isbn-13 978-0-521-83951-8 hardback
isbn-10 0-521-83951-3 hardback

Cambridge University Press has no responsibility for the persistence or accuracy of
urls for external or third-party internet websites referred to in this book, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.

Contributors page vii
Preface xi

1 Introduction 1
John Christman and Joel Anderson

part i the self: conceptions of the
autonomous self
2 Decentralizing Autonomy: Five Faces of Selfhood 27
Diana Tietjens Meyers
3 The Self as Narrator 56
J. David Velleman
4 Autonomy and Self-Identity 77
Marina A. L. Oshana

part ii the interpersonal: personal authority
and interpersonal recognition
5 Taking Ownership: Authority and Voice in
Autonomous Agency 101
Paul Benson
6 Autonomy, Vulnerability, Recognition, and Justice 127
Joel Anderson and Axel Honneth
7 Autonomy and Male Dominance 150
Marilyn Friedman


part iii the social: public policy and liberal
8 Autonomy, Domination, and the Republican Challenge to
Liberalism 177
Richard Dagger
9 Liberal Autonomy and Consumer Sovereignty 204
Joseph Heath
10 Political Liberty: Integrating Five Conceptions of
Autonomy 226
Rainer Forst

part iv the political: liberalism, legitimacy,
and public reason
11 Liberalism without Agreement: Political Autonomy and
Agonistic Citizenship 245
Bert van den Brink
12 The Place of Autonomy within Liberalism 272
Gerald F. Gaus
13 Moral Autonomy and Personal Autonomy 307
Jeremy Waldron
14 Autonomy, Self-Knowledge, and Liberal Legitimacy 330
John Christman

Bibliography 359
Index 377

Joel Anderson is Research Lecturer in Philosophy at Utrecht University
(The Netherlands). He works on issues of personal autonomy, practi-
cal reasoning, neuro-ethics, mutual recognition, and moral psychology.
He has published articles in various journals, including Philosophical Ex-
plorations, Constellations, Deutsche Zeitschrift f¨ r Philosophie, and Philosophy,
Psychiatry, and Psychology.
Paul Benson is Professor and Chair of the Philosophy Department at
the University of Dayton. He works in the areas of ethics, action theory,
and social philosophy. He has published articles on autonomy, oppressive
socialization, and self-worth. He is completing a book tentatively entitled
The Place of Self-Worth in Free Agency.
Bert van den Brink is Research Lecturer in Philosophy at Utrecht Uni-
versity (The Netherlands), specializing in contemporary social and polit-
ical philosophy. He is the author of The Tragedy of Liberalism: An Alterna-
tive Defense of a Political Tradition (SUNY Press, 2000) and co-editor, with
Maureen Sie and Marc Slors, of Reasons of One™s Own (Ashgate, 2004).
John Christman is Associate Professor of Philosophy at Pennsylvania State
University. He has written on property rights, individual autonomy, and
liberal political philosophy. He is the author of The Myth of Property: To-
ward an Egalitarian Theory of Ownership (Oxford University Press, 1994)
and Social and Political Philosophy: A Contemporary Introduction (Routledge,
2002) and is the editor of The Inner Citadel: Essays on Individual Autonomy
(Oxford University Press, 1989).


Richard Dagger is Professor of Political Science and Philosophy at Ari-
zona State University, where he directs the Philosophy, Politics, and Law
Program for Barrett Honors College. He works in the areas of rights,
political obligation, punishment, and other topics in political and legal
philosophy. His books include Civic Virtues: Rights, Citizenship, and Repub-
lican Liberalism (Oxford University Press, 1997) and, with Terence Ball,
Political Ideologies and the Democratic Ideal (Longmans, 2004).
Rainer Forst is Professor of Political Theory in the Departments of So-
cial Sciences and of Philosophy, J. W. Goethe University (Frankfurt, Ger-
many). His areas of specialization are political philosophy and ethical
theory. He is the author of Contexts of Justice: Political Philosophy Beyond
Liberalism and Communitarianism (University of California Press, 2002)
and Toleranz im Kon¬‚ikt: Geschichte, Gehalt und Gegenwart eines umstrittenen
Begriffs (Suhrkamp, 2003).
Marilyn Friedman is Professor of Philosophy at Washington University in
St. Louis and works in the areas of ethics, feminist theory, and political
philosophy. Her books include Autonomy, Gender, Politics (Oxford Univer-
sity Press, 2002) and What Are Friends For?: Feminist Perspectives on Personal
Relationships and Moral Theory (Cornell University Press, 1993).
Gerald F. Gaus is Professor of Philosophy at Tulane University, New Or-
leans. He is a faculty member of the Murphy Institute of Political Econ-
omy. His research interests are in political philosophy, social philosophy,
and ethics. His books include Justi¬catory Liberalism (Oxford University
Press, 1996), Value and Justi¬cation: The Foundations of Liberal Theory (Cam-
bridge University Press, 1990), and Contemporary Theories of Liberalism:
Public Reason as a Post-Enlightenment Project (Sage, 2003). He is a founding
co-editor of the journal Politics, Philosophy, and Economics.
Joseph Heath is Associate Professor of Philosophy at the University of
Toronto. He writes on political theory, moral philosophy, and rational
choice theory. He is the author of Communicative Action and Rational Choice
(MIT Press, 2001), The Ef¬cient Society (Penguin, 2001), and, with Andrew
Potter, The Rebel Sell (Harper Collins, 2004).
Axel Honneth is Professor of Social Philosophy at J. W. Goethe University
in Frankfurt, Germany, and Director of the Institute for Social Research
there. He has published on issues of political philosophy, ethics, moral
psychology, and social theory. His books in English include Critique of
Power (MIT Press, 1991), The Struggle for Recognition: The Moral Grammar
Contributors ix

of Social Con¬‚icts (Polity Press, 1995), The Fragmented-World of the Social
(SUNY Press, 1995), The Morality of Recognition (Polity Press, 2004), and,
with Nancy Fraser, Recognition or Redistribution?: A Political-Philosophical
Exchange (Verso, 2003).
Diana Tietjens Meyers is Professor of Philosophy at the University of
Connecticut, Storrs. In the spring of 2003, she was awarded the Blanche,
Edith, and Irving Laurie New Jersey Chair in the Women™s and Gender
Studies Department at Rutgers University. Her most recent monographs
are Subjection and Subjectivity: Psychoanalytic Feminism and Moral Philoso-
phy (1994) and Gender in the Mirror: Cultural Imagery and Women™s Agency
(2002). A collection of her (mostly) previously published articles, Being
Yourself: Essays on Identity, Action, and Social Life, appeared in 2004. She is
the editor of Feminists Rethink the Self and Feminist Social Thought: A Reader.
She is the author of the forthcoming Encyclopedia Britannica article on
philosophical feminism.
Marina A. L. Oshana is Associate Professor of Philosophy at the Univer-
sity of Florida. Her research focuses on issues in normative ethics and
moral psychology. She has published articles in the area of autonomy
and responsibility and is completing a book tentatively entitled Personal
Autonomy: Its Breadth and Limits.
J. David Velleman is G. E. M. Anscombe Collegiate Professor of Philosophy
at the University of Michigan, specializing in philosophy of action, ethics,
and philosophy of mind. He is the author of Practical Re¬‚ection (Princeton
University Press, 1989), The Possibility of Practical Reason (Oxford Univer-
sity Press, 2000), and Self to Self (Cambridge University Press, forthcom-
ing). He is a founding co-editor of the journal Philosophers™ Imprint.
Jeremy Waldron is the Maurice and Hilda Friedman Professor of Law at
Columbia University, where he also is Director of the Center for Law and
Philosophy. He has published widely in legal and political theory. He is
the author of God, Locke and Equality (Cambridge University Press, 2002);
The Dignity of Legislation (Cambridge University Press, 1999); Law and Dis-
agreement (Oxford University Press, 1999); and Liberal Rights (Cambridge
University Press, 1993).

The initial idea for this volume was to prepare an update of The Inner
Citadel, the collection of essays on the concept of autonomy that John
Christman had put together in 1989. Given the spate of terri¬c work
since then, a new anthology seemed in order. But we also saw that dis-
cussions of the concept of autonomy needed to engage more fully with
the growing body of literature on political liberalism, where there were
strikingly similar lines of critique and rebuttal. Thus arose the idea for a
collection of essays that would both update discussions of autonomy and
connect them to debates over the foundations of liberalism.
The decision to solicit new essays allowed us to tailor our invitations
to authors in a way that framed these issues from the outset, and we
are particularly pleased with the way the authors took up and further
developed those issues. The chapters were all written independently, but
during the process of revising their contributions, the authors had access
to drafts of each other™s chapters, which allowed for interesting cross-
pollination and a more cohesive overall volume. In addition, several of
the authors had an earlier opportunity to exchange their views at symposia
on autonomy in St. Louis in 1997 and 1999.
We would like to acknowledge Sigur°ur Kristinsson for his part in or-
ganizing these symposia with Joel Anderson, and Washington University
in St. Louis and the University of Missouri, St. Louis, for supporting the
As with any complex, collaborative project, putting together this vol-
ume has required hard work and patience by many people “ most impor-
tantly, our contributors. We are very appreciative of their commitment to
the project. At Cambridge University Press, Terence Moore™s faith in the

project allowed it to get off the ground in the ¬rst place. Ronald Cohen
handled the manuscript editing in an ef¬cient and thorough manner
and gave valuable stylistic advice. And Daniel Brunson helped greatly
with the index and the ¬nal preparation of the manuscript. We thank all
these individuals for their efforts.
We would also like to thank our respective academic departments for
support for research connected with this project: for John Christman, the
Departments of Philosophy and Political Science at Penn State, and for
Joel Anderson, the departments of Philosophy at Washington University
in St. Louis and, subsequently, at Utrecht University.
Finally, two personal notes. John would like to express his love and
gratitude to Mary Beth Oliver for her insights, patience, and support
throughout the process. And Joel would like to thank Pauline Kleingeld
for helping to make it all not only possible but also so much better.
Autonomy and the Challenges to Liberalism
New Essays


John Christman and Joel Anderson

Recent theoretical debates over political liberalism address a wide variety
of issues, from citizenship and minority rights to the role of constitu-
tional foundations and democratic deliberation. At stake in virtually all
of these discussions, however, is the nature of the autonomous agent,
whose perspective and interests are fundamental for the derivation of
liberal principles. The autonomous citizen acts as a model for the basic
interests protected by liberal principles of justice as well as the repre-
sentative rational agent whose hypothetical or actual choices serve to
legitimize those principles. Whether implicitly or explicitly, then, cru-
cial questions raised about the acceptability of the liberal project hinge
on questions about the meaning and representative authority of the au-
tonomous agent. Similarly, in the extensive recent philosophical litera-
ture on the nature of autonomy, debates over the content-neutrality of
autonomy or the social conditions necessary for its exercise ultimately
turn on issues of the scope of privacy, the nature of rights, the scope of
our obligation to others, claims to welfare, and so on “ the very issues
that are at the heart of discussions of liberalism regarding the legitimate
political, social, and legal order.
Despite the conceptual and practical interdependence of liberalism
and autonomy, however, the recent literature on liberalism has devel-
oped without much engagement with the parallel boom in philosophical
work on autonomy, and vice versa. This book serves as a point of intersec-
tion for these parallel paths. The chapters connect the lines of inquiry
centering on the concept of autonomy and the self found in relatively less
“political” areas of thought with the debates over the plausibility of lib-
eralism that have dominated political philosophy in the Euro-American
John Christman and Joel Anderson

tradition for some time. While the main focus of the collection is to
explore the intersection we are describing, the chapters also represent
efforts to make free-standing contributions to debates about autonomy
as well as to the foundations and operations of liberal justice itself.
In what follows, we begin by outlining the recent debates over auton-
omy, before noting some of the challenges to liberalism that have mo-
tivated current rethinking within political theory. We then discuss four
key themes at issue in both the debates over autonomy and the debates
over liberalism: value neutrality, justi¬catory regresses, the role of inte-
gration and agreement, and the value of individualism. This is followed,
by a summary of each of the chapters, with a brief discussion of how the
individual essays create a dialogue among themselves concerning these
broad and fundamental issues of political philosophy.

I An Initial Characterization of Autonomy
As we map the terrain of these controversies, it will be helpful to spell
out the central features of the conception of autonomy, and some key
distinctions relating to it, that predominate in discussions of autonomy
and autonomy-based liberalism.
Three terminological distinctions are central here. First is that be-
tween moral and personal autonomy. “Moral autonomy” refers to the ca-
pacity to subject oneself to (objective) moral principles. Following Kant,
“giving the law to oneself” in this way represents the fundamental or-
ganizing principle of all morality.1 “Personal autonomy,” by contrast, is
meant as a morally neutral (or allegedly neutral) trait that individuals
can exhibit relative to any aspects of their lives, not limited to questions
of moral obligation.2 Under some understandings of the term, for ex-
ample, one can exhibit personal autonomy but reject or ignore various
of one™s moral obligations. The chapters by Forst (10), Gaus (12), and
Waldron (13) speci¬cally address this distinction.3 Second, the autonomy
of persons can, in principle, be separated from local autonomy “ autonomy
relative to particular aspects of the person, say, her desires. Though the
question of whether these ideas can and should be separated is an issue
that theorists have directly debated in the literature.4 Finally, we can dis-
tinguish between “basic” autonomy “ a certain level of self-government
necessary to secure one™s status as a moral agent or political subject “
and “ideal” autonomy “ the level or kind of self-direction that serves as a
regulative idea but not (or not necessarily) a set of requirements we must
meet to secure our rights, be held morally responsible, and enjoy other
status designators that basic autonomy mobilizes.
Introduction 3

These distinctions are important, but the notion of autonomy still
¬nds its core meaning in the idea of being one™s own person, directed
by considerations, desires, conditions, and characteristics that are not
simply imposed externally on one, but are part of what can somehow be
considered one™s authentic self.5 There is disagreement about whether
the concept should rest on reference to a “true” self (see, for example,
the chapters in Part I), but in general the focus is on the person™s compe-
tent self-direction free of manipulative and “external” forces “ in a word,
To govern oneself, one must be in a position to act competently and
from desires (values, conditions, and so on) that are in some sense one™s
own.6 This delineates the two families of conditions that have played cen-
tral roles in recent debates over autonomy: authenticity conditions and com-
petency conditions. Authenticity conditions are typically built on the capac-
ity to re¬‚ect on and endorse (or identify with) one™s desires, values, and
so on. The most in¬‚uential model “ that developed by Gerald Dworkin
and Harry Frankfurt7 “ views autonomy as requiring second-order iden-
ti¬cation with ¬rst-order desires. Competency conditions specify that
agents must have various capacities for rational thought, self-control,
self-understanding, and so on “ and that they must be free to exercise
those capacities, without internal or external coercion.8 Dworkin sums
up this hierarchical account by saying that autonomy involves second-
order identi¬cation with ¬rst-order desires under conditions of “proce-
dural independence” “ that is, conditions under which the higher-order
identi¬cation was not in¬‚uenced by processes that subvert re¬‚ective and
critical capacities.9
This standard conception of autonomy ¬ts well with standard accounts
of political liberalism “ and not by accident. In particular, the notion of
“procedural independence” is meant to specify in a non-substantive way
the conditions under which individual choice would count as authori-
tative “ that is, in a way that makes no reference to constraints on the
content of a person™s choices or the reasons he or she has for them. In
a thoroughly liberal manner, this shift to formal, procedural conditions
allows this model to accommodate a diversity of desires and ways of life
as autonomous.

II Challenges to Liberalism™s Reliance on the
Autonomous Individual
Within recent discussions of liberalism, debates over the nature of au-
tonomy have emerged from a slightly different viewpoint. Liberalism can
John Christman and Joel Anderson

be characterized in a number of ways, a point addressed in several of
the chapters here, but it generally involves the approach to the justi¬-
cation of political power emerging from the social contract tradition of
the European Enlightenment, where the authority of the state is seen to
rest exclusively on the will of a free and independent citizenry.10 Justice,
de¬ned with reference to basic freedoms and rights, is thought to be
realized in constitutional structures that constrain the individual and col-
lective pursuit of the good. Central to the speci¬cation of justice in this
tradition are the interests and choices of the independent, self-governing
citizen, whose voice lends legitimacy to the power structures that enact
and constitute justice in this sense.11
The multivocal contestation of this tradition has often centered on
the conception of the person that functions as both sovereign and sub-
ject of principles of justice. In particular, the conception of the per-
son as an autonomous, self-determining and independent agent has
come under ¬re from various sources. Communitarians and defend-
ers of identity politics point to the hyper-individualism of such a view “
the manner in which the autonomous person is seen as existing prior
to the formulation of ends and identities that constitute her value ori-
entation and identity. Feminists point up the gender bias implicit in
the valorization of the independent “man” devoid of family ties and
caring relations; communitarians note the inability of such a view to
make full sense of the social embeddedness of persons; and various
postmodernists decry assumptions of a stable and transparent “self”
whose rational choices, guided by objective principles of morality, de-
¬ne autonomous agency. From these various directions, the model
of the autonomous person has drawn powerful calls for reconsideration.
What has emerged from recent discussions of both liberalism and the
nature of the autonomous self is a set of controversies that mirror each
other in provocative and constructive ways. Amidst the wide range of such
controversies, four stand out as particularly relevant for our purposes: the
question of value-neutrality, the problem of foundations, the question-
able emphasis placed on unity and agreement, and the allegedly hyper-
individualism of both autonomy-based liberalism and standard accounts
of the autonomous self.

IIa Value Neutrality
One of the major disagreements in the philosophical literature is over
whether autonomy should be understood in a “procedural” “ and hence
Introduction 5

“value-neutral” “ manner, or whether it is better understood in a “substan-
tive” way. The latter view is defended for example, by Marina Oshana and
Paul Benson in their chapters (4 and 5). On this view, autonomy must
include conditions that refer to substantive value commitments, both by
the autonomous person herself and by those around her “ conditions
concerning her own self-worth, the constraints others set, and the like.
A driving force behind the call for substantive conceptions is, among
other things, the claim that autonomy should not be seen as compat-
ible with certain constrained life situations “ such as positions of social
domination and self-abnegation “ no matter how “voluntarily” the person
came to choose or accept that situation.12
Correspondingly, critics of liberalism have claimed that “procedural”
liberalism fails to take account of the way in which fundamental value
commitments constitute the identities and motivational structures of
those citizens expected to accept and endorse principles of justice.13
Like the defenders of substantive accounts of autonomy, “perfectionist”
critics of liberalism claim that mechanisms of liberal legitimacy cannot
demand of citizens that they bracket from deliberation of political prin-
ciples those commitments that constitute their very identities.14 These
critics charge that “neutralist” liberalism removes from the political pro-
cess the motivational anchor of these deep commitments, without which
it is dif¬cult to stave off political apathy and maintain civic engagement.15
And strict value-neutrality requirements even threaten to “gag” citizens
from expressing their most heartfelt concerns within the political process.
With regard to both autonomy and liberalism, then, critics have raised
the question of how one can ground political legitimacy in a conception
of autonomous choice without allowing substantive values (communitar-
ian or perfectionist) to play some role in the conception of autonomy

IIb The Regress Problem and the Foundations of Liberal Legitimacy
In another complex discussion concerning the conceptual conditions of
autonomy, the issue has been raised as to whether re¬‚ective endorsement
of ¬rst-order desires (or other aspects of the personality) is necessary or
suf¬cient for the authenticity required of autonomy. Commentators have
pointed out that such a condition invites a regress, since the question is
left open as to whether any given act of endorsement (and the desires
and values it rests on) merits the authenticity that it itself bestows on ¬rst-
order aspects of the self. If so, and if authenticity is established through
John Christman and Joel Anderson

critical re¬‚ection, then a third-order desire must be postulated to ground
an endorsement of the second-order desire in order to retain the ¬rst.
But this merely raises the same question once again concerning that
third-order desire, and so on. Yet, if even the second-order appraisal is
not tested for its authenticity, the question is left open as to whether a
person thoroughly manipulated in her desires and values (hypnotized,
brain-washed, etc.) would be called autonomous if those second-order
attitudes were themselves manipulated by her captors.16
Critics of “hierarchicalist” conceptions of autonomy have also raised
the question of why intrasubjective endorsement confers normative au-
thority on ¬rst-order wants and values in the ¬rst place. What is special
about the higher-order voices that render other aspects of the self so
(metaphysically) special? We can certainly imagine cases where a per-
son™s ¬rst-order drives and motives are better re¬‚ections of their inde-
pendent and self-governing natures (their “true selves,” if you wish) than
second-order re¬‚ections, which may themselves simply mirror relentless
conditioning and inauthentic responses to social pressures. This point
is touched on in the chapters by Meyers (2), Benson (5), and Christ-
man (14). Meyers and Benson both express skepticism, for example, that
higher-order re¬‚ective endorsement is the core element of autonomy in
all its important guises, while Christman claims that in the context of lib-
eral political theory, seeing autonomy as including self-re¬‚ection of this
sort is crucial, despite dif¬culties with that process.17
In the political realm, a similar issue arises with regard to the tra-
ditional liberal assumption that citizens™ choice is suf¬cient to legit-
imize political principles and policies. Critics have long been skeptical
of the claim that mere public acclamation of some issue, even if such
approval has been re¬‚ected on and consciously endorsed with reasons,
re¬‚ects unmanipulated and independent voices when there exists perva-
sive ideological and other social pressures working to undermine such
independent re¬‚ection.18 These discussions parallel questions about a
regress of conditions for autonomy in asking whether political legitimacy
requires something more than the collective endorsement of political
preferences. Similarly, it can be asked of procedural liberalism why
plebescitary endorsement by legislative bodies (the element of govern-
ment corresponding to “higher-order” re¬‚ection) should automatically
render the judgments they produce legitimate. One of the challenges
that democratic liberalism has always faced stems from cases in which
formally valid procedures lead to abhorrent results, results that may
Introduction 7

even threaten the very foundations of liberalism. Is democracy its own
justi¬cation, or must there be “extra-legislative” constitutional checks
to ensure free, independent debate in the public sphere and ground

IIc The Problematic Emphasis on Integration, Unity, and Agreement
Whereas the previous two challenges to standard approaches to auton-
omy and liberalism suggest the need for a more substantive approach,
two other lines of critique accuse such approaches of unduly substantive
(and contestable) value commitments. These critics charge that standard
accounts of autonomy and liberalism are less value-neutral and pluralist
than they claim, for they actually presuppose, for example, values of per-
sonal integration, or egoistic individualism. And the problems this raises
concern not only theoretical coherence but also the inclusiveness of so-
cial and political application of principles centering on autonomy so
Various writers focusing on the standard conception of the au-
tonomous person have raised trenchant questions about the degree
to which such conceptions problematically assume a uni¬ed, self-
transparent consciousness lurking in all of us and representing our most
settled selves. These commentators point to the ways in which con¬‚ict
and irresolvable ambivalence characterize the modern personality. They
emphasize that our motivational lives must be understood as containing
various elements that are hidden from re¬‚ective view and disguised or
distorted in consciousness (as Meyers, and Anderson and Honneth, dis-
cuss in their chapters, 2 and 6). The idea of uni¬ed, transparent selves
being a mark of autonomy has thus come to be seen as suspect.
In a parallel manner, critical analyses of political liberalism have cen-
tered on the desirability and coherence of demanding full collective en-
dorsement by the governed in order to establish legitimacy. As van den
Brink (11) suggests in his chapter, liberalism without agreement may well
suit the deep and abiding con¬‚icts (as well as multiple identities) charac-
teristic of modern societies. Additionally, there has been much discussion
among (especially) Marxist and other radical writers of the way in which
liberalism™s pretensions of deliberative transparency ignore or suppress
what truly drives the social and political movements in a society “ the
dynamics of economic and social power and its often hierarchical distri-
bution and exercise.20
John Christman and Joel Anderson

IId Individualism
Also prominent in recent literature on both autonomy and liberalism are
discussions of the alleged hyper-individualism of the liberal conception
of the autonomous person. Feminists have developed extensive critiques
of the overly masculine emphasis on separated, atomistic decisions op-
erating in this conception. Communitarians have famously claimed that
the liberal emphasis on autonomy has obscured the socially embedded
nature of identity and value.21 Motivated by these and related critiques,
calls have been made to recon¬gure the idea of autonomy in ways that
take more direct account of the social nature of the self and the relational
dynamics that de¬ne the value structure of most people. “Relational” and
“social” accounts of autonomy have been developed to respond to such
calls, de¬ning the autonomous person in ways that make direct reference
to the social components of our identities and value commitments.22
The chapters by Meyers (2), Benson (5), Oshana (4), and Anderson and
Honneth (6) all touch on this issue.
Communitarians, feminists, defenders of identity politics, and others
have long claimed that liberal political philosophy rests on an unaccept-
ably individualist understanding of human value and choice.23 Some
liberal theorists have insisted that the charge of hyper-individualism is
overdrawn.24 Others, famously, have followed Rawls™s “political” turn in
claiming that models of personhood at work in political principles serve
merely a representative function for the purposes of consensus and com-
promise, rather than claiming universalistic applicability or metaphysical
truth.25 But other theorists have taken a second look at the idea of person-
hood at the center of liberalism, and adopted more socially embedded
conceptions meant to be sensitive to charges of exclusionary individu-
alism of this sort.26 However, in the chapters by Dagger (8), Forst (10),
Heath (9), and Anderson and Honneth (6), the issue of the split between
traditional liberal individualism and more social conceptions of the self
(as, for example, in “republican” traditions) is examined in a manner
that sheds new light on these con¬‚icts.
As can be seen from this review of these four broad challenges, there
are parallel implications for discussions of the conceptual structure of au-
tonomy and for debates over the problems and promise of liberal political
philosophy. There is thus much to be gained by bringing these discus-
sions together. The chapters collected here represent just this kind of
cross-pollenation. Although the discussions of liberalism and autonomy
are interwoven throughout, we have arranged them thematically in a pro-
gression of sorts, tracing a spiral that moves from conceptions of the self
Introduction 9

and the individual (where autonomy has been conceptualized in seem-
ingly less “political” ways) to the confrontation between self and other,
to the role of autonomy in evaluative interpretations of social life and
social policies, and then ¬nally to the overt consideration of the political-
theoretical importance of autonomy in the foundations of liberalism.

III The Self: Conceptions of the Autonomous Self (Part I)
Since liberalism is centrally a view about the extent of legitimate inter-
ference with the wishes of the individual, it is not surprising that debates
over liberalism have centered on the nature of the self. The respect that
individuals claim for their preferences, commitments, goals, projects, de-
sires, aspirations, and so on is ultimately to be grounded in their being
the person™s own. It is because those preferences, commitments, and so
on are a person™s own that disregarding them amounts to disregarding
him or her qua that distinctive individual. By contrast, disregarding pref-
erences, commitments, and so on that are the product of coercion or
deception does not seem to involve a violation in the same sense, raising
the vexing issue of what makes some preferences, commitments, and so
on “one™s own,” and others not. Given the recent pressure on concepts of
the true self, authenticity, or re¬‚ectively endorsed higher-order desires,
further work is needed in order to clarify the grounds for treating indi-
viduals as the autonomous agents of their lives or the sovereign source of
political authority. Central to this work are the questions “ regarding the
nature of the self “ taken up in Part I by Diana Tietjens Meyers, David
Velleman, and Marina Oshana.
In her chapter (2), “Decentralizing Autonomy: Five Faces of Selfhood,”
Meyers challenges the standard liberal assumption that autonomy is ex-
clusively a matter of re¬‚ective self-de¬nition and rational integration.
She develops an account of autonomous agency as a matter of navigat-
ing a complex plurality of demands. Most fundamentally, she argues for
the need to redress many theorists™ overemphasis on self-de¬nition to
the neglect of self-discovery. Whereas self-de¬nition is a matter of the self-
analysis and inner endorsement so prominent in hierarchical accounts,
self-discovery is more diffuse, and more a matter of sensitivity and open-
ness. In order to clarify the skills needed for self-discovery “ and to un-
derscore their importance “ Meyers develops a “¬ve-dimensional account
of the self”: the self as unitary, social, relational, divided, and embodied.
Corresponding to each of these dimensions of the self, she suggests, are
agentic skills that are crucial to autonomy. Capacities for critical re¬‚ection
John Christman and Joel Anderson

and ego-integration are among them, but they belong to only one of
the registers in which we come to discover who we are or even exer-
cise self-direction. For, as Meyers points out, autonomy often emerges in
unexpected places: the unexpected smashing of dishes in the sink, the
body™s refusal to relinquish its hold on life, or even a revealing slip of the
tongue. Meyers concludes that unless we have the skills to stay in touch
with the non-unitary and non-individual components of the self, we lack
what is needed for full autonomy, however good we might be at critical
Like Meyers, Velleman (Chapter 3) is concerned with the issue of how
to understand autonomous agency once one has given up the idea that
there is a “true self” to be discovered. If the self turns out not to be a ¬xed
star to guide one™s deliberations but rather a shifting, inchoate, plural,
and perhaps even illusory point of reference, it becomes much harder
to say what it is that makes some desires truly one™s own and others not.
Unlike Meyers, Velleman does see uni¬cation of the self as a central com-
ponent of autonomous agency. Taking as his point of departure Daniel
Dennett™s idea that the self is no more real than a person™s center of grav-
ity “ that the self is simply one™s “narrative center of gravity” “ Velleman
argues that although our selves are indeed our narrative inventions, they
are nonetheless real, because “we really are the characters whom we in-
vent.” Velleman is not, however, defending the view that anything goes,
that there are no constraints on that narrative. But neither are these con-
straints external to the self. His ingenious move here is to point out that we
not only identify narrative patterns in our actions, we also choose actions
so as to ensure that there is a pattern into which they will ¬t. Otherwise,
we cannot make sense of ourselves. The idea of the self as narrator is thus
not a fantasy of arbitrary control; we cannot make ourselves up simply by
wishing. Instead, when we are living the life we are narrating, it is built
into the task that we have to ensure both that the narrative ¬ts the life
and that the life continues to ¬t the narrative. This does not require that
autonomous agents always continue a past trajectory, but any departures
from past patterns must then cohere with a larger narrative identity and
But however much we may write our own narratives, we do so under
conditions that are not of our own choosing. This is a central theme in
Oshana™s chapter (4). She takes up the thorny issue of whether “ and,
if so, under what conditions “ one can act autonomously on the basis of
inescapable components of one™s identity. Classical liberal conceptions
of autonomy typically focus on voluntary consent as the sole basis for
Introduction 11

legitimate choice, whether in the domain of personal autonomy or polit-
ical deliberation. This suggests that one acts autonomously only if one acts
from values, desires, traits, and so on that one could give up if one wanted
to. In the 1980s, this assumption of detachment came under ¬re from
such theorists as Harry Frankfurt and Michael Sandel, who argued that
such a requirement would eliminate far too many of our best reasons for
acting. In particular, if “sheddability” were a necessary condition for a
component of one™s identity to count as a grounds for autonomous
action, then it would be non-autonomous to act out of love for family
members or, in general, from many of our deepest commitments (com-
mitments, incidentally, that liberalism was designed to protect).28 But if
autonomy involves acting from reasons that are most fully one™s own, then
it would seem that conceptions of autonomy must not rule out attach-
ments and commitments, for it is often precisely those that it is unthinkable
for us to give up that are most centrally constitutive of who we are.29 As
Oshana points out, however, some de¬ning and inescapable components
of one™s identity may be unwanted. She insightfully analyzes her own case
of having ascribed to her the racial identity of an African-American. This
racial attribution is inescapable and clearly determinative of who she
is, despite the fact that, as a biracial woman, she is alienated from it.
This seems to generate an unwelcome implication for authenticity-based
accounts of autonomy. For if autonomy requires wholehearted endorse-
ment of one™s self-conception, then one cannot allow into one™s self-
conception any components about which one is ambivalent. But in some
cases, Oshana argues, this creates an indefensible disjunction between
either being autonomous or viewing oneself clearly “ for example, ac-
knowledging the social reality of being African-American. One response,
for which Oshana has a great deal of sympathy, is to say that this is a
further cost of living in a racist society, and that promoting autonomy
is a matter of promoting justice, racial and otherwise. Her core theo-
retical response, however, is to call for a rethinking of the requirement
that one not be alienated from components of one™s identity. It may be,
she suggests, that full authenticity is not actually necessary for autonomy.

IV The Interpersonal: Personal Authority and Interpersonal
Recognition (Part II)
Oshana™s point about the ambivalent character of having one™s identity
tied to the attitudes of others provides a bridge to the chapters in Part II,
which situate the exercise of autonomy within the interpersonal domain.
John Christman and Joel Anderson

The chapters by Paul Benson (5) Joel Anderson and Axel Honneth (6),
and Marilyn Friedman (7), represent distinct approaches to the idea of
the “social,” “relational,” or “intersubjective” self that emerged especially
in feminist work of the late-1980s, and has continued since.30 A central
challenge faced by defenders of “social” conceptions of autonomy is how
to acknowledge the ways in which individuals™ most authentic desires are
not merely generated within but even authorized by their social context,
while at the same time keeping in mind the ways in which interpersonal re-
lations can distort and dominate individuals™ desires. There is widespread
agreement on rejecting the idea that authenticity and autonomy come
exclusively through retreating into an “inner citadel” of detached, higher-
order re¬‚ection. What is less clear, however, is what should replace this
notion, if one is to avoid eviscerating the idea that exercising autonomy “
and demanding the respect for individual autonomy that is central to
liberalism “ is a matter of opposing others™ demands for conformity or
Benson™s approach to autonomy focuses on the dual aspects of being
accountable to others for one™s self-authorization. In light of various dif¬-
culties with accounts of autonomy that focus on identifying with one™s
motivational states, Benson argues that we should rethink the active, re-
¬‚exive character of autonomy in terms of the agent™s assertion of her
authority to speak for her actions, and the desires, values, and so on that
provide the warrant for those actions. As autonomous agents, we invest
ourselves in our actions by vouching for ourselves as authorized to speak
for them. This emphasis on autonomous agents™ re¬‚exive attitudes is in
line with standard views of autonomy. But Benson™s approach departs
from such views in analyzing autonomous actions as the actions of agents
who vouch for their authority to give reasons for their behavior, should
they be called on to do so. This shift to a social and discursive perspec-
tive on self-authorization raises anew the issue so central to liberalism “
how to understand the authority of those whose accounts of their ac-
tions are dismissed in the larger social context. The stakes are high here,
for as Benson points out, “internalized invisibility can defeat agents™ ca-
pacities to take ownership of what they do.” One central dif¬culty with
approaches that conceptualize autonomy as requiring that one be able to
actually satisfy others™ demands for an account of oneself is that it can end
up denying marginalized voices the authority (as autonomous agents) to
assert their concerns. Benson™s strategy for avoiding this dif¬culty is to tie
autonomy not to a stronger requirement of full discursive competence
but rather to the act of taking responsibility for responding to “potential
Introduction 13

challenges which, from [a person™s] own point of view, others might ap-
propriately bring to his view.” In this way, Benson™s account situates the
idea of personal autonomy within a social and discursive context, but still
leaves the focus on the claim that individuals stake to being heard.
Like Benson, Anderson and Honneth situate autonomy within the
interpersonal context of answering for one™s actions, and they too are
concerned with the ways in which a lack of social recognition can im-
pair an individual™s autonomy. But their view of recognition differs from
that of Benson. For Anderson and Honneth, autonomy emerges only
within “ and is largely constituted by “ relations of mutual recognition.31
Building on that central idea, they focus on the vulnerabilities of indi-
viduals regarding the development and maintenance of their autonomy
and, in particular, on the question of what it would mean for a society to
take seriously the obligation to minimize individuals™ autonomy-related
vulnerabilities. According to their dialogical model of autonomy, indi-
viduals are much more deeply dependent on their social environment
for the acquisition, maintenance, and exercise of their autonomy than
liberals usually acknowledge. Therefore, questions of social justice need
to be reframed to focus on equality of access to participation in the re-
lations of recognition through which individuals acquire the autonomy
needed for true freedom.
Friedman is also deeply concerned about the ways in which rela-
tionships of inequality, injustice, and domination undermine personal
autonomy, and especially the autonomy of women in interpersonal re-
lationships. But her approach is quite different from Anderson and
Honneth™s. Although she would no doubt agree that more just and egal-
itarian social relations would greatly enhance the opportunities for de-
veloping personal autonomy, instead she reframes the question of the
nature and value of autonomy in terms of the question of what auton-
omy women need in the face of apparently intransigent patterns of male
domination. Given the evidence that male domination is likely to be a
long-term feature of the social world, she argues, the type of autonomy
particularly valuable to women is the capacity to resist subordination, by
“acting for the sake of wants or desires that were not adapted to mimic
the wants or needs of their dominators.” Thus, although Friedman is
well aware of the importance of social relationships for the formation
of autonomous selfhood,32 her research into the dynamics of domestic
violence in particular has lead her to sound a clear note of caution regard-
ing attempts to rethink autonomy in ways that make it indistinguishable
from oppressive forms of accommodation and submission. As her chapter
John Christman and Joel Anderson

here makes clear, however “social” or “relational” autonomy may be, what
lends such urgency to its value is its role in shielding individuals within
relational and political contexts from oppression and subordination.

V The Social: Public Policy and Liberal Principles (Part III)
One traditional way of drawing the line between liberal and “republican”
approaches to political principles is in terms of the level of public partic-
ipation and active citizenship required by one™s status as a free person.
Traditionally, the liberal emphasis on the “liberty of the moderns” has
placed protection from social and political pressures to engage in public
activity at the center of conceptions of justice, whereas republican politics
have linked the obligations of public life and participation in the collec-
tive self-government that de¬nes social freedom with the status of a free
citizen. Richard Dagger, in his chapter (8), argues that when autonomy is
seen as occupying a central place in (the best version of) republican pol-
itics, this contrast blurs signi¬cantly. He makes this point by examining
two in¬‚uential recent attempts to revitalize republican theory (by Philip
Pettit and Quentin Skinner), attempts that seek to highlight the contrast
with liberalism. Dagger shows how close attention to the concept of au-
tonomy relied on in both traditions dilutes the supposed clash between
these traditions.
Many of the alleged tensions between liberalism and traditional re-
publican conceptions of justice also turn on the contested meaning
of political freedom or liberty and its relationship to an understanding
of citizen autonomy, especially insofar as that understanding assumes
a division (and potential opposition) between autonomy as individu-
alized self-government and autonomy as collective, socially instituted
self-legislation. The complex relationship between individual liberty and
autonomy (in both its individualized as well as more social manifesta-
tions) has been the subject of numerous discussions.33
Rainer Forst (Chapter 10) investigates the meaning of political liberty
in a way that rests on, and insists on the protection of, ¬ve conceptions
of autonomy, each of them salient in different contexts but all related to
the overall protection of citizen sovereignty (in both the individual and
collective senses). To enjoy political liberty, for Forst, is to enjoy the sta-
tus of the citizen of a political community and thereby to be positioned
to engage in procedures of reciprocal justi¬cation of guiding principles.
This “intersubjective” conception of liberty depends on the protection of
Introduction 15

individual autonomy in various respects, including: (1) moral autonomy
(the ability to act on reasons that take others into account, and thereby
contribute to the justi¬cation of coercive practices); (2) ethical autonomy
(the formation of a distinct identity and conception of the good, in-
cluding second-order abilities to re¬‚ect on and alter such conceptions);
(3) legal autonomy (the protection from being forced to live according to
others™ value conceptions); (4) political autonomy (maintaining one™s sta-
tus as a participant in public justi¬cation); and (5) social autonomy (hav-
ing access to the internal and external means of securing one™s status as
a member of the political community). Forst concludes that “citizens are
politically free to the extent to which they, as freedom-givers and freedom-
users, are morally, ethically, legally, politically, and socially autonomous
members of a political community.”
The challenges to classical liberalism coming from republican political
theory highlight the precariousness of assumptions about citizens™ ability
to choose independently of social pressures.34 The lively debate around
such issues of political sovereignty contrasts sharply with discussions of
“consumer sovereignty,” which tends to be either quietistic and uncriti-
cal or naive and paternalistic. The familiar challenge is how to accom-
modate strong intuitions about the way in which “consumerist” pressures
(say, from advertising) lead to substantively bad choices without slipping
into paternalism or elitism about people™s choices. In his chapter (9),
Joseph Heath argues that such a critique is possible, but that theorists
must proceed with caution. After identifying the failures in typical cri-
tiques of consumer sovereignty, Heath argues that two such lines of cri-
tique “ those focusing on failures of collective action “ can indeed be
defended, but these critical strategies should be understood as resting on
a richer understanding and appreciation of consumer autonomy rather
than on a call for its limitation. In the process, Heath highlights the need
for a more nuanced conception of individual autonomy. For example, he
mentions that critiques of consumer sovereignty based on the pervasive
nature of advertising often naively consign all socially in¬‚uenced desires
to the category of non-autonomy, critiques that display an over-zealous
skepticism about preference change, thereby masking a surreptitious per-
fectionism of political values. However, a fully worked out model of au-
tonomy that is meant to apply to contexts of this sort should aid us in
differentiating socially manipulated desires changes from merely socially
in¬‚uenced ones, and of doing so in a way that does not end up violating
principles of value-neutrality. In this way, critical appraisals of the role of
John Christman and Joel Anderson

advertising in undermining autonomous consumer choice would rest on
¬rmer ground.

VI The Political: Liberalism, Legitimacy,
and Public Reason (Part IV)
As mentioned earlier, autonomy ¬gures in the structure of liberalism as
the feature of the subject whose endorsement of principles of justice pro-
vide the fundamental legitimacy of those principles. However, assuming
full consensus “ such as a Rawlsian overlapping consensus by agents who
consider themselves free and equal autonomous persons but who are
motivated by mutually incompatible moral viewpoints “ is not universally
accepted as either a necessary or a suf¬cient condition for the justi¬ca-
tion of principles of justice. Bert van den Brink argues in his chapter (11),
in fact, that liberalism without agreement must be accepted as the work-
ing model of justi¬cation in light of the deep and abiding multiplicity
of value frameworks in the modern world. Establishing legitimacy with-
out agreement demands a collective understanding of a constitutional
structure that is itself evolving and subject to review. The autonomous
citizen under such a model is more than merely the bearer of rights;
she is a person with the capacity both to accept and to contest concep-
tions of citizenship on which such constitutional structures rest. Public
reason, then, demands that citizens be secured not only capacities for
deliberation and public discourse, but also the social virtues of “civic en-
durance” and “civic responsiveness.” The former must be exercised by
those victims of social inequality who contest dominant principles but
who must accept the evolving nature of social institutions. The latter is
required of those (alleged) bene¬ciaries of social injustice who must be
open to challenges from victims and sensitive to the systematic ways in
which such challenges can be suppressed and misunderstood. The result,
van den Brink argues, is the establishment of “agonistic” autonomy, as a
component of reasonable pluralism.
Liberalism is often characterized as based on the commitment of the
priority of the right over the good. However, an alternative view of liberal-
ism sees its foundations as more substantive “ namely, that the particular
principle of “right” that must be secured prior to the promotion of the
good of citizens is the protection of individual liberty.35 On this view, lib-
eralism is a political morality that requires that any interference with
the freedom of action of any person is unjust unless that interference
Introduction 17

can be justi¬ed, and it must be justi¬ed in terms that the victim of the
interference can somehow accept as a reason. This ties the structure of
liberalism inherently to the giving of reasons and the justi¬cation of ac-
tions. Gerald Gaus, in his chapter (12), explores this structure and its
implications for the conception of autonomy that functions at its center.
Gaus points out how liberalism is traditionally understood to rest on
the value of personal autonomy, autonomy conceived in a morally neutral
manner without speci¬c reference to substantive values. Moral auton-
omy, on the other hand, takes up the Kantian mantle of de¬ning the
self-governing person as having the capacity to grasp certain objective
moral norms. Gaus argues, however, that insofar as liberalism requires
that interferences be justi¬ed on the basis of reasons that all accept “ and
the standard for acceptance displays a modest internalism by claiming
that such reasons must appeal to considerations operative in or accessi-
ble by the motivational system of the person accepting the reason “ then
liberalism cannot rest simply on the protection of personal autonomy.
For unless we understand the autonomy of citizens as containing com-
mitments to shared moral norms, then no such general justi¬cations can
be successful, and the overall legitimacy of coercive political principles
(all of which involve interferences with freedom of action) would be lost.
This, then, is Gaus™s way of addressing the issues of value-neutrality and
justi¬catory regress raised earlier.
The line between personal and moral autonomy, and that distinc-
tion™s relevance to liberal political theory, is the central target of Jeremy
Waldron™s chapter (13). Waldron examines Kant™s positions on the im-
portance of protecting individual freedom and, in particular, Kant™s claim
that pursuing one™s own happiness is morally praiseworthy, even though
it involves the heteronomous pursuit of one™s own desires, whereas be-
ing coerced by another is categorically wrong, even though it equally
involves being moved by external desires “ that is, desires that are not
fully one™s own (in Kant™s strict sense). Indeed, one can wonder, Waldron
suggests, whether Kantian theory makes room for valuing personal au-
tonomy as such. There is some basis for interpreting Kant as seeing rea-
son as playing a key role in the choice of non-moral ends, and hence
securing a basis for respect of others™ pursuit of happiness (despite its
involvement in pathological desire). But the question remains as to what
extent moral autonomy (the value-laden, substantive conception of self-
government) is implicated in the traditional liberal respect for (only)
personal autonomy?
John Christman and Joel Anderson

The central liberal principle that citizens should be allowed to pursue
their own conception of the good involves recognition of personal au-
tonomy insofar as that pursuit is understood to proceed autonomously “
that is, as the pursuit of ends endorsed by second-order (or some such)
re¬‚ection and evaluation. Moreover, in a political conception of liber-
alism, such as Rawls™s, there must also be general consensus on such
principles achieved in a way consistent with each seeing herself and her
co-citizens as free and equal persons autonomously pursuing a plan of
life. (This “seeing” of herself and others need not involve believing it
to be true of them, merely that they can be represented as such for the
purposes of consensus building.) For the overlapping consensus to be
generated, however, people must be willing to circumscribe their con-
ception of the good by the conditions necessary for a similar pursuit on
the part of others. And given the deep plurality of such conceptions, the
value of personal autonomy must be kept clearly separate from the value
of moral autonomy, since the latter de¬nes autonomy with reference to
a single comprehensive set of moral values. But personally autonomous
citizens do not merely endorse their ¬rst-order preferences out of some
passing desire; rather they see their individual commitments as an act
of conscience “ a morally obligatory commitment to self-imposed princi-
ples (manifesting, that is, moral autonomy). So the problem is that if
personal autonomy and moral autonomy are seen as too separate, it is
unclear why personally autonomous citizens following their conscience
would be willing to circumscribe their pursuits by the requirements of
consensus. But if the autonomy respected in the liberal state is moral
autonomy, then respect for a deep and abiding plurality of moral view-
points is thereby threatened. Waldron leaves us with that ponderous and
trenchant dilemma.
In the ¬nal chapter (14), John Christman takes up the role of auton-
omy in public reason and liberal legitimacy. He confronts, in particular,
those critics who argue that autonomy-based liberalism is problematic be-
cause it makes unreasonable assumptions about a person™s ability to know
herself. Christman begins by clarifying and supplementing the claim that
persons are systematically opaque to themselves regarding their motiva-
tions, deepest commitments, and psychological dynamics. Despite this,
however, respecting people™s autonomy in ways that ask them to repre-
sent themselves, so to speak, is required by the dynamics of collective
choice and public reason that political legitimacy depends on. Public
reason is necessary for the establishment or even merely the aim of legit-
imating principles of justice, and the dynamics of public reason demand
Introduction 19

that participants engage with each other as sincere representatives of
points of view who are willing to give reasons to others as a way of jus-
tifying (potentially) shared principles, and to do this in a way that does
not revert simply to a Hobbesean clash of desires. Seeing the process of
public legitimation this way provides a principled argument for recog-
nizing and respecting people™s abilities to re¬‚ectively endorse their own
commitments (their autonomy) despite the admission that in doing so
we will often systematically misunderstand our own deepest motives. But
holding people responsible for what they re¬‚ectively accept about them-
selves is essential in the dynamic of democratic interchange that political
legitimacy demands.

VII Conclusion
This attention to the relationship between different conceptions of au-
tonomy and the requirements of public deliberation brings to the fore
a set of themes that weave through virtually all of the papers in this vol-
ume. In what ways can autonomy be de¬ned so as to take seriously the
broad multiplicity of value orientations, modes of reasoning and re¬‚ec-
tion, conceptions of identity, and approaches to politics and social life
that mark the modern condition? And how can respect for autonomy
take seriously the way that identities as well as abilities to pursue values
and relationships are fundamentally structured by the social dynamics
one ¬nds oneself within, social dynamics whose very structure ought to
be the subject of politics? The key tensions in debates over the meaning of
autonomy “ substantive versus procedural notions, the contested require-
ment of re¬‚ective self-endorsement, the complex relationship between
internal authenticity and social de¬nitions of identity, and so on “ are
replicated in political debates over the possibility of legitimate principles
of justice in a complex, pluralistic world. What these chapters at least
show is the irresponsibility, if not impossibility, of separating these lines
of inquiry: the conceptual, the moral, and the political are all mutually
implicated in re¬‚ections on these issues.

Thomas Hill, “The Kantian Conception of Autonomy,” in John Christman,
ed., The Inner Citadel: Essays on Individual Autonomy (New York: Oxford
University Press, 1989), 91“105.
Gerald Dworkin, The Theory and Practice of Autonomy (New York: Cambridge
University Press, 1988), 34“47.
John Christman and Joel Anderson

3. According to Waldron, for example, the standard Kantian distinction be-
tween the pursuit of mere desire “ what personal autonomy refers to “ and
willful adoption of the moral law “ moral autonomy “ breaks down once one
considers the role these notions play in the protection of individual liberty
that is central to liberal politics.
4. See, for example, Marina Oshana, “Personal Autonomy and Society,” The
Journal of Social Philosophy 29 ( 1998): 81“102.
5. For a general discussion of the concept of autonomy, see Bernard
Berofsky, Liberation from Self. New York: Cambridge University Press,
1995; John Christman, entry on “Autonomy in Moral and Political Phi-
losophy” in Edward Zalta, ed., The Stanford Encyclopedia of Philosophy,
http://plato.stanford.edu/contents.html; Gerald Dworkin, The Theory and
Practice of Autonomy. New York: Cambridge University Press, 1988; Lawrence
Haworth, Autonomy: An Essay in Philosophical Psychology and Ethics. New Haven:
Yale University Press, 1986; Richard Lindley, Autonomy. Atlantic Highlands,
NJ: Humanities Press International, 1986; Catriona Mackenzie and Natalie
Stoljar, “Introduction: Autonomy Re¬gured,” in Relational Autonomy: Feminist
Perspectives on Autonomy, Agency, and the Social Self. New York: Oxford Univer-
sity Press, 2000; Alfred Mele, Autonomous Agents: From Self-Control to Autonomy.
New York: Oxford University Press, 1995; and Diana T. Meyers, Self, Society,
and Personal Choice. New York: Columbia University Press, 1989.
6. As has regularly been pointed out, conceptions of autonomy that see only
desires as the focal point will be too narrow, as people can exhibit autonomy
relative to a wide variety of personal characteristics, such as values, physical
traits, relations to others, and so on; any element of body, personality, or
circumstance that ¬gures centrally in re¬‚ection and action should be open
to appraisal in terms of autonomy (see Richard Double, “Two Types of Au-
tonomy Accounts,” Canadian Journal of Philosophy 22 (1992): 65“80, p. 66).
7. Frankfurt™s view is not explicitly an account of autonomy, but rather of free-
dom of the will. See Harry Frankfurt, “Freedom of the Will and the Concept
of a Person,” in The Importance of What We Care About (Cambridge: Cambridge
University Press, 1987), 80“94. Nevertheless, the account has been absorbed
into the literature on autonomy as a model of that notion.
8. On the variety of competency conditions, see for example, Bernard Berofsky,
Liberation from Self, Chapter 8; Robert Young, Autonomy: Beyond Negative and
Positive Liberty (New York: St. Martin™s Press, 1986); Lawrence Haworth, Au-
tonomy, 83“122; and Diana T. Meyers, Self, Society, and Personal Choice, 76“97.
9. See Gerald Dworkin, “The Concept of Autonomy,” in John Christman, ed.,
The Inner Citadel: Essays on Individual Autonomy (New York: Oxford University
Press, 1989), 54“62, p. 61. Partially in response to objections of the sort
discussed, Dworkin has revised his view to exclude an explicit requirement
of identi¬cation. For Dworkin, autonomy involves (among other things)
the capacity to raise the question of whether one identi¬es with the desires in
question (The Theory and Practice of Autonomy, p. 15).
10. Of course, this model of popular sovereignty has been known to exhibit
serious exclusionary tendencies, speci¬cally concerning the makeup of this
“citizenry.” For a discussion, see Carol Pateman The Sexual Contract (Stanford,
Introduction 21

CA: Stanford University Press, 1988) and Charles Mills, The Racial Contract
(Ithaca, NY: Cornell University Press, 1997).
As noted, this characterization is the subject of much debate. For overview
discussions, see Will Kymlicka, Liberalism, Community and Culture (Oxford:
Clarendon, 1989), chapters 2“3, and John Christman, Social and Political
Philosophy: A Contemporary Introduction (London: Routledge, 2002), Part I.
See, for example, Natalie Stoljar, “Autonomy and the Feminist Intuition,” in
Relational Autonomy; Paul Benson, “Freedom and Value” Journal of Philosophy
84 ( 1987): 465“86; Sigurdur Kristinsson, “The Limits of Neutrality: Toward
a Weakly Substantive Account of Autonomy,” Canadian Journal of Philosophy 30
( 2000): 257“86; Diana T. Meyers, “Feminism and Women™s Autonomy: The
Challenge of Female Genital Cutting,” Metaphilosophy Vol. 31 no. 5 (October,
2000): 469“91; and Robert Paul Wolff, In Defense of Anarchism (New York:
Harper & Row, 1970).
See Michael Sandel, Liberalism and the Limits of Justice, 2nd ed. (Cambridge:
Cambridge University Press, 1999).
See Joseph Raz, The Morality of Freedom (Oxford: Clarendon, 1986); George
Sher, Beyond Neutrality: Perfectionism and Politics (Cambridge: Cambridge Uni-
versity Press, 1997); Thomas Hurka, Perfectionism (New York: Oxford Univer-
sity Press, 1993); Bert van den Brink, The Tragedy of Liberalism: An Alterna-
tive Defense of a Political Tradition (Albany, NY: SUNY Press, 2000); Richard
Dagger, Civic Virtues: Rights, Citizenship, and Republican Liberalism (Oxford:
Oxford University Press, 1997); and Steven Wall, Liberalism, Perfectionism and
Restraint (New York: Cambridge University Press, 1998).
See Charles Taylor, “What™s Wrong with Negative Liberty?” in Philosophy and
the Human Sciences, vol. 2 of Philosophical Papers (Cambridge: Cambridge Uni-
versity Press, 1985), 211“29.
For discussion of this dif¬culty, see Christman, “Introduction” to The Inner
Citadel (New York: Oxford University Press, 1989), 11.
Earlier discussion of this issue can be found in Marilyn Friedman, “Autonomy
and the Split-Level Self,” Southern Journal of Philosophy, vol. 24, no. 1 ( 1986)
19“35, and Irving Thalberg, “Hierarchical Analyses of Unfree Action,” in The
Inner Citadel, 123“36. In fact, Frankfurt himself acknowledges that “The mere
fact that one desire occupies a higher level than another in the hierarchy
seems plainly insuf¬cient to endow it with greater authority or with any
constitutive legitimacy” (Frankfurt, “Identi¬cation and Wholeheartedness,”
in The Importance of What We Care About, 166f.). Indeed, this is already clear in
the 1976 essay, “Identi¬cation and Externality,” reprinted in The Importance
of What We Care About ; see esp., 65f.
This charge takes many forms, from focusing on how advertising subverts
supposedly free-market behavior to the manner in which public interchange
has become subverted by the dominance of corporate and ideological struc-
turing. See, for example, J¨ urgen Habermas, The Structural Transformation of
the Public Sphere (Cambridge, MA: M.I.T. Press, 1991).
This is a complex issue about which much has been written. See, for exam-
ple, Jurgen Habermas, Between Facts and Norms, William Rehg, trans. (Cam-
bridge, MA: MIT Press, 1996); and Jeremy Waldron, The Dignity of Legislation
John Christman and Joel Anderson

(Cambridge: Cambridge University Press, 1999), for discussion of some as-
pects of it.
See, for example, Michel Foucault Discipline and Punish: The Birth of the
Prison, Alan Sheridan, trans. (New York: Vintage Books, 1995); David Harvey,
The Condition of Postmodernity (Oxford: Basil Blackwell, 1989); William Con-
nolly, Identity/Difference (Ithaca, NY: Cornell University Press, 1991); and
Wendy Brown, States of Injury: Power and Freedom in Late Modernity (Princeton,
NJ: Princeton University Press, 1995). Liberal writers who show sensitiv-
ity to these challenges (in some form) include Donald Moon, Constructing
Community: Moral Pluralism and Tragic Con¬‚icts (Princeton, NJ: Princeton
University Press, 1993) and Bert van den Brink, The Tragedy of Liberalism. For
an overview discussion, see John Christman, Social and Political Philosophy: A
Contemporary Introduction, chapter 7.
See, for example, Carol Gilligan, In a Different Voice: Psychological Theory and
Women™s Development (Cambridge, MA: Harvard University Press), 1982; Nell
Noddings, Caring: A Feminist Approach to Ethics and Moral Education (Berkeley,
CA: University of California Press, 1984); Michael Sandel, Liberalism and the
Limits of Justice ; and Alasdair MacIntyre, After Virtue (Notre Dame, IN: Notre
Dame University Press, 1984).
See, for example, Catriona Mackenzie and Natalie Stoljar, “Autonomy Re-
con¬gured” and John Christman “Feminism and Autonomy,” in Dana
Bushnell, ed., Nagging Questions (Lauham, MD: Rowman & Little¬eld, 1994),
See, for example, the essays in Mackenzie and Stoljar, eds., Relational Auton-
omy and Young, Justice and the Politics of Difference, chapter 4).
See Will Kymlicka, Liberalism, Community and Culture, 47“99; see also Forst,
Contexts of Justice: Political Philosophy Beyond Liberalism and Communitarian-
ism, John M. M. Farrell, trans. (Berkeley, CA: University of California Press,
Rawls, Political Liberalism (New York: Columbia University Press, 1993); John
Gray, Post-Liberalism: Studies in Political Thought (New York: Routledge, 1993).
See, for example, Jack Crittenden, Beyond Individualism (New York: Oxford
University Press, 1992).
For a discussion that raises questions about the use of narrativity as a condi-
tion of personhood, see John Christman, “Narrative Unity as a Condition of
Personhood,” Metaphilosophy (Oct. 2004).
For a discussion of this issue, see Alfred Mele, Autonomous Agents, Chapter. 4;
John Christman, “Liberalism, Autonomy, and Self-Transformation” Social
Theory and Practice Vol. 27, no. 2 ( 2001), 185“206; and Joel Anderson,
“Autonomy and the Authority of Personal Commitments: From Internal
Coherence to Social Normativity,” Philosophical Explorations: An International
Journal for the Philosophy of Mind and Action 6 ( 2003): 90“108.
See Harry Frankfurt, The Importance of What We Care About: Chapter 13;
Necessity, Volition, and Love (Cambridge: Cambridge University Press, 1987):
Chapters 9, 11, 14, and Sandel, Liberalism and the Limits of Justice.
See Diana T. Meyers, Self, Society, and Personal Choice; Seyla Benhabib, Situat-
ing the Self: Gender, Community and Postmodernism in Contemporary Ethics (New
Introduction 23

York: Routledge 1992); Marilyn Friedman, What are Friends For? Feminist Per-
spectives on Personal Relationships and Moral Theory (Ithaca, NY: Cornell Uni-
versity Press, 1993); and “Autonomy and Social Relationships: Rethinking
the Feminist Critique” in D. Meyers, ed., Feminists Rethink the Self (Boulder,
CO: Westview Pres, 1997), 40“61; Paul Benson, “Feminist Second Thoughts
About Free Agency” Hypatia ( 1990): 47“64; John Christman, “Feminism
and Autonomy”; Eva F. Kittay, Love™s Labor: Essays on Women, Equality and De-
pendency (New York: Routledge, 1999); Loraine Code, “Second Persons,” in
What Can She Know? Feminist Theory and the Construction of Knowledge (Lanham,
MD: Rowman and Little¬eld, 1991). See also Jurgen Habermas, “Individua-
tion through Socialization: On George Herbert Mead™s Theory of Subjectiv-
ity,” in Postmetaphysical Thinking, William Hohengarten, trans. (Cambridge,
Mass.: MIT Press, 1992), 149“204; and Joel Anderson, “W¨ nsche zweiter
Ordnung, starke Wertungen und intersubjektive Kritik: Zum Begriff ethis-
che Autonomie.” Deutsche Zeitschrift f¨ r Philosophie 42 ( 1994): 97“119.
See Honneth, The Struggle for Recognition: The Moral Grammar of Social Con¬‚icts,
Joel Anderson, trans. (Cambridge: Polity Press, 1995).
See Friedman, What Are Friends For? (Cornell University Press, 1993) and
Autonomy, Gender and Politics (Oxford: Oxford University Press, 2003).
Most notably, perhaps, Isaiah Berlin, “Two Concepts of Liberty,” in Four Essays
on Liberty (London: Oxford University Press, 1969), 118“72.
This is particularly clear in Philip Pettit™s “republican” account of freedom
as non-domination in A Theory of Freedom.
See, for example, Raymond Geuss, “Liberalism and Its Discontents” Political
Theory, Vol. 30, no. 3 ( 2001): 320“39.
part i


Conceptions of the Autonomous Self

Decentralizing Autonomy

Five Faces of Selfhood

Diana Tietjens Meyers

People are cast into highly variable and unpredictable circumstances.
Sometimes they face appalling situations. Sometimes they face predica-
ments of mind-boggling complexity or paralyzing opacity. Even the most
familiar, seemingly routine situations are nuanced in unforeseen ways,
and ignoring these subtleties can only lead to missteps, misunderstand-
ings, or worse. I take it that an account of autonomy should capture the
agentic resourcefulness people need to cope with life™s vicissitudes, or-
deals, and upheavals.1 To do this, an account of autonomy must explain
how one can encounter unexpected constraints, discern novel opportu-
nities, and improvise on the spot without parting company from one™s
authentic traits, affects, values, and desires. More speci¬cally, a tenable
account of self-discovery and self-de¬nition must be premised on a view
of authenticity that countenances suf¬cient adaptability to make sense
of these agentic capacities. In this chapter, I seek to extend the range of
autonomous agency while preserving a rich enough view of autonomous
re¬‚ection and choice to draw the vital distinction between enacting au-
thentic attributes and enacting inauthentic ones.2
There are all sorts of good reasons to classify conduct as nonau-
tonomous, but I suspect that philosophers misclassify some conduct be-
cause it stems from agentic capacities that have wrongly fallen into dis-
repute among autonomy theorists. Autonomy theorists for whom Kant™s
moral philosophy is the locus classicus tend to gravitate to a mentalis-
tic, individualistic conception of the autonomous subject and to a ratio-
nalistic account of autonomous deliberation and volition. In this view,
forms of agency that evidently are not anchored in rational powers are
deemed autonomous only if they can somehow be assimilated to reason.
Diana Tietjens Meyers

Consequently, much of the philosophical literature is devoted to design-
ing rational certi¬cation procedures to draw conduct into the orbit of
Two considerations have led me to question autonomy theory™s focus
on critical reason and rationally mandated volition. First, there are several
creditable conceptions of the self in widespread use both in scholarly
contexts and in everyday discourse, and it strikes me as troubling that
the idea of autonomy has become so entwined with one of them that the
others seem altogether problematic from the standpoint of autonomy.
To ¬‚esh out this concern, I set out ¬ve conceptions of the self “ the
unitary self, the social self, the relational self, the divided self, and the
embodied self (Section I). For each conception, I sketch how it represents
the constitution of individual identity, and I explain how that view of
identity sets up friction with autonomy. Second, thinking freshly about
my own experience has led me to suspect the privileging of one of these
conceptions “ the unitary self “ over the others. Bringing two recent
experiences (recounted in Section II) to bear on this issue prompted
me to reconsider the role of the relational self and the embodied self
in autonomy. The kinds of experiences I describe present puzzles for
autonomy theory because, although it is hard to believe that my conduct
did not comport with authentic traits, affects, values, and desires, it is far
from clear that I rationally reviewed and endorsed these attributes, and
it is all too clear that my will was not under rational control.
In my view, the best way to meet the challenge posed by these experi-
ences is to recognize the social self, the relational self, the divided self,
and the embodied self as potential sites of autonomous self-discovery,
self-de¬nition, and self-direction. I begin to make my case for this claim
by describing forms of practical intelligence associated with each of these
conceptions of the self (Section III). But since agreeing that there are
agentic skills linked to each conception of the self does not rule out deny-
ing that these skills secure autonomy, I consider how a theory of retro-
spective autonomy or a “personal style” theory of autonomy might explain
the autonomy of the sorts of experience I sketch in Section II without
adverting to any of these skills (Section IV). I argue, however, that neither
type of theory provides a convincing analysis of the forms of agency that
interest me, although they do point up the need to rethink self-discovery
and self-de¬nition. Thus, my strategy is to focus on self-discovery and self-
de¬nition and to argue that a plausible account of these processes would
accommodate the agentic skills of the social self, the relational self, the
divided self, and the embodied self, as well as the unitary self (Section V).
Decentralizing Autonomy: Five Faces of Selfhood 29

But an obvious objection to this decentralized approach is that I have
fractured autonomous subjectivity beyond repair “ that ¬ve-dimensional
subjectivity is an unwieldy, disjointed monstrosity. In reply, I explain how
easily and cogently autobiographical narratives reconcile these seemingly
disparate motifs (Section VI). Since this self-descriptive form is available,
there is no need to reduce autonomy to its rationalistic dimension, and
autonomy theory can make sense of otherwise unintelligible autonomy

I Five Conceptions of the Self and Five-Dimensional Subjectivity
In this section, I lay out what I take to be the ¬ve principal conceptions
of the self that are commonly invoked in vernacular discourses and that
are currently prominent in philosophical thinking, as well. They are the
unitary self, the social self, the relational self, the divided self, and the
embodied self. Associated with each of these conceptions is a distinctive
endowment of desirable attributes and capacities. Likewise, each con-
ception provides a particular kind of answer to the question of what an
individual is like “ that is, each sees individual identity as derived from

. 1
( 13)